JZUS - Journal of Zhejiang University SCIENCE

Frontiers of Information Technology & Electronic Engineering

Accepted manuscript available online (unedited version)

Efficient learning of robust multigait quadruped locomotion for minimizing cost of transport

Author(s): Zhicheng WANG¹, Xin ZHAO², Meng Yee (Michael) CHUAH³, Zhibin LI⁴, Jun WU^1, ⁵, Qiuguo ZHU^1, ⁵
Affiliation(s): ¹Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou 310000, China; more
Corresponding email(s): pekingbit@163.com
Key Words: Reinforcement learning; Locomotion; Motor learning; Energy efficiency

Share this article to： More <<< Previous Paper \|Next Paper >>>

Zhicheng WANG¹, Xin ZHAO², Meng Yee (Michael) CHUAH³, Zhibin LI⁴, Jun WU^1,⁵, Qiuguo ZHU^1,⁵. Efficient learning of robust multigait quadruped locomotion for minimizing cost of transport[J]. Frontiers of Information Technology & Electronic Engineering,in press.https://doi.org/10.1631/FITEE.2401070

@article{title="Efficient learning of robust multigait quadruped locomotion for minimizing cost of transport",
author="Zhicheng WANG¹, Xin ZHAO², Meng Yee (Michael) CHUAH³, Zhibin LI⁴, Jun WU^1,⁵, Qiuguo ZHU^1,⁵",
journal="Frontiers of Information Technology & Electronic Engineering",
year="in press",
publisher="Zhejiang University Press & Springer",
doi="https://doi.org/10.1631/FITEE.2401070"
}

%0 Journal Article
%T Efficient learning of robust multigait quadruped locomotion for minimizing cost of transport
%A Zhicheng WANG¹
%A Xin ZHAO²
%A Meng Yee (Michael) CHUAH³
%A Zhibin LI⁴
%A Jun WU^1
%A⁵
%A Qiuguo ZHU^1
%A⁵
%J Frontiers of Information Technology & Electronic Engineering
%P
%@ 2095-9184
%D in press
%I Zhejiang University Press & Springer
doi="https://doi.org/10.1631/FITEE.2401070"

TY - JOUR
T1 - Efficient learning of robust multigait quadruped locomotion for minimizing cost of transport
A1 - Zhicheng WANG¹
A1 - Xin ZHAO²
A1 - Meng Yee (Michael) CHUAH³
A1 - Zhibin LI⁴
A1 - Jun WU^{1
A1 -}⁵
A1 - Qiuguo ZHU^{1
A1 -}⁵
J0 - Frontiers of Information Technology & Electronic Engineering
SP -
EP -
%@ 2095-9184
Y1 - in press
PB - Zhejiang University Press & Springer
ER -
doi="https://doi.org/10.1631/FITEE.2401070"

Abstract
Chinese Summary
Academic Network
Reviewer Comment

Abstract: Quadrupedal robots are able to exhibit a range of gaits, each with its own traversability and energyefficiency characteristics. By actively coordinating between gaits in different scenarios, energy-efficient and adaptivelocomotion can be achieved. This study investigates the performance of learned energy-efficiency policies forquadrupedal gaits under different commands. We propose a training-synthesizing framework that integrates learnedgait-conditioned locomotion policies into an efficient multiskill locomotion policy. The resulting control policiesachieved low-cost smooth switching and controllable gaits. Our results of the learned multiskill policy demonstrateseamless gait transitions while maintaining energy optimality across all commands

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

Open peer comments: Debate/Discuss/Question/Opinion

<1>

- Go to

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference